Representing Procedural Knowledge for Semantic Networks using Neural Nets
Reinhard Moratz, Gunther Heidemann, Stefan Posch, Helge Joachim Ritter, Gerhard Sagerer · 1995
Interpretation of complex scenes involves analysing multiple objects being composed of several parts. Since different objects often have parts in common it is useful to share resources representing these identical substructures. This is quite a difficult task for artificial neural networks (ANNs) [4], but can be handled with semantic networks. They are a well established tool for representing and organizing scene models [1, 7]. However, signal interpretation needs to be robust against distortions and adaptive in different environments. These properties are characteristic advantages of ANNs [12] which suggests to combine them with semantic nets. In this paper we propose a combination of ANNs and semantic networks attempting to combine the benefits of both approaches. First results on visual recognition of composed objects are promising. 1 Introduction Describing a given scene is the purpose of computer vision systems. Classification of simple objects from extracted features is feasable...